At physiological pH, many drugs (eg. penicillins, NSAIDs, methotrexate, thiazides) exist as organic anions and are eliminated via the kidney proximal tubule through multispecific renal organic anion transporters (especially OAT1 and OAT3). Apart from their general relevance to renal drug elimination, OATs are of current interest because they mediate nephrotoxicity of the antivirals cidofovir and adefovir, drugs which could be used in the event of a smallpox epidemic. Due to their toxicity, treatment and prophylaxis of many individuals with these drugs (as might be required in an epidemic) is likely to result in significant morbidity. Here, we hypothesize that altered function and expression of OAT1 and OAT3 due to coding and noncodinq region SNPs are key determinants of individual variation in renal drug elimination and nephrotoxicitv from drugs like cidofovir. The Pi's lab is in an ideal position to address this question as it was the first to identify the prototypical member of this family (originally called NKT, later termed OAT1), as well as several other SLC22 family members, and the first to describe a knockout of an OAT (OAT3). As part of this REVISED proposal, we aim to: 1) identify human SNPs in OAT1 and OATS in ethnically diverse samples;2) employ two independent computational approaches (structural modeling and phylogenetic footprinting) to prioritize SNPs in terms of likelihood of functional effects on drug transport and OAT gene expression;3) analyze the effects of coding region SNPs in order of in silica priority on in vitro drug transport (eg. using Xenopus oocyte and COS-7 expression systems) and the effects of noncoding SNPs on OAT gene expression (eg.by performing transfections of promoter constructs). We provide evidence of our ability to do all these studies and also have well-established (published) collaborations with individuals who have particular expertise in techniques that will support the project. We have done our best to address all the concerns of the previous review. Considerable new preliminary data, including major refinements in the computational model, are presented. Since a large, healthy, phenotyped group of patients treated with nephrotoxic drugs like cidofovir does not exist, we argue that this computational strategy of prioritizing OAT SNPs, followed by in vitro analysis of function and expression, is a logical approach for understanding the impact of OAT1 and OAT3 polymorphisms on handling of drugs and their toxicity. Ultimately, this work should provide the basis of a useful strategy to diminish renal failure in humans treated with nephrotoxic drugs.